Research On Control Technology By Tactile Brain Computer Interface

The brain-computer interface is a new type of human-computer interaction technology that converts brain activity signals into control commands for manipulation of external devices.This paper considers the necessity of constructing a visual independent brain-computer interface paradigm in some specific scenarios,selects the tactile channel as the research object,and combines the motion imaging task to design the hybrid paradigm and the corresponding robot control system.Firstly,based on the existing tactile P300 paradigm,the robot arm control system is designed.The subjects select the corresponding action of the robot by paying attention to the vibration stimulation on the arm,and finally complete the grasping task.In the online control experiment,the average time spent on the five subjects was 182.7 seconds,which took 12.9 commands.Compared with other control paradigms,the haptic paradigm can control the multi-degree-of-freedom controlled object without occupying the visual channel,but it also has the problem of low control efficiency.Then the paper designs a hybrid brain-computer interface paradigm based on the tactile P300 task and the motion imaging task.In the training experiment,the performance of the experiment is adjusted due to the poor performance of the three-class sports imaging task.Imagine that ”left hand movement” and ”right hand movement” are unified into the ”imagination” state,thus obtaining a better training accuracy rate.In the subsequent online test experiments,the subjects achieved an accuracy of more than 70Based on the hybrid paradigm,the robot arm control system was designed by using the hybrid paradigm.By completing the same grab task,the difference between the actual control and the single-mode tactile P300 paradigm was compared.And the number of commands generated is reduced by about 50Finally,in view of the fatigue and state decline of the subjects during the training of the sports imagination task,it is explored whether the commonly used transcranial electrical stimulation method in medicine will affect the EEG signal and improve the fatigue state.The results show that for most of the subjects,the classifier obtained by training the EEG signals of normal state and fatigue state can effectively distinguish the EEG signals of the post-stimulation state and the fatigue state.This result has certain enlightening significance for the subsequent research.